Improving Graph Few-shot Learning with Hyperbolic Space and Denoising Diffusion

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Liu, Yonghao, Sun, Jialu, Pang, Wei, Giunchiglia, Fausto, Li, Ximing, Feng, Xiaoyue, Guan, Renchu
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918475420663808
author Liu, Yonghao
Sun, Jialu
Pang, Wei
Giunchiglia, Fausto
Li, Ximing
Feng, Xiaoyue
Guan, Renchu
author_facet Liu, Yonghao
Sun, Jialu
Pang, Wei
Giunchiglia, Fausto
Li, Ximing
Feng, Xiaoyue
Guan, Renchu
contents Graph few-shot learning, which focuses on effectively learning from only a small number of labeled nodes to quickly adapt to new tasks, has garnered significant research attention. Despite recent advances in graph few-shot learning that have demonstrated promising performance, existing methods still suffer from several key limitations. First, during the meta-training phase, these methods typically perform node representation learning in Euclidean space, which often fails to capture the inherently hierarchical structure existing in real-world graph data. Second, during the meta-testing phase, they usually fit an empirical target distribution derived from only a few support samples, even when this distribution significantly deviates from the true underlying distribution. To address these issues, we propose IMPRESS, a novel framework that IMproves graPh few-shot learning with hypeRbolic spacE and denoiSing diffuSion. Specifically, our model learns node representations in a hyperbolic space and enriches the support distribution through denoising diffusion mechanisms. Theoretically, IMPRESS achieves a tighter generalization bound. Empirically, IMPRESS consistently outperforms competitive baselines across multiple benchmark datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2604_27462
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Improving Graph Few-shot Learning with Hyperbolic Space and Denoising Diffusion
Liu, Yonghao
Sun, Jialu
Pang, Wei
Giunchiglia, Fausto
Li, Ximing
Feng, Xiaoyue
Guan, Renchu
Machine Learning
Artificial Intelligence
Graph few-shot learning, which focuses on effectively learning from only a small number of labeled nodes to quickly adapt to new tasks, has garnered significant research attention. Despite recent advances in graph few-shot learning that have demonstrated promising performance, existing methods still suffer from several key limitations. First, during the meta-training phase, these methods typically perform node representation learning in Euclidean space, which often fails to capture the inherently hierarchical structure existing in real-world graph data. Second, during the meta-testing phase, they usually fit an empirical target distribution derived from only a few support samples, even when this distribution significantly deviates from the true underlying distribution. To address these issues, we propose IMPRESS, a novel framework that IMproves graPh few-shot learning with hypeRbolic spacE and denoiSing diffuSion. Specifically, our model learns node representations in a hyperbolic space and enriches the support distribution through denoising diffusion mechanisms. Theoretically, IMPRESS achieves a tighter generalization bound. Empirically, IMPRESS consistently outperforms competitive baselines across multiple benchmark datasets.
title Improving Graph Few-shot Learning with Hyperbolic Space and Denoising Diffusion
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2604.27462